EI2GYB > ASTRO 18.11.25 13:15l 53 Lines 4997 Bytes #200 (0) @ WW
BID : 47707_EI2GYB
Read: GUEST
Subj: Machine learning discovers quasars acting as lenses
Path: ED1ZAC<ED1ZAC<IZ3LSV<IK6IHL<IK7NXU<HB9ON<DK0WUE<PI8ZTM<VK2IO<VE3CGR<
K1AJD<GB7BED<EI2GYB
Sent: 251118/1404Z 47707@EI2GYB.DGL.IRL.EURO LinBPQ6.0.25
_ _
(_\ /_)
)) ((
.-"""""""-.
_ _ _ _ /^\/ _. _. \/^\
/ \ ___| |_ _ __ ___ | \ | | _____ _____ \( /__\ /__\ )/
/ _ \ / __| __| '__/ _ \ | \| |/ _ \ \ /\ / / __| \, \o_/_\o_/ ,/
/ ___ \\__ \ |_| | | (_) | | |\ | __/\ V V /\__ \ \ (_) /
/_/ \_\___/\__|_| \___/ |_| \_|\___| \_/\_/ |___/ `-.'==='.-'
__) - (__
+------------------------------------------------------------------------------+
Machine learning discovers quasars acting as lenses
Quasars acting as strong gravitational lenses are among the rarest finds in astronomy. Out of nearly 300,000 quasars cataloged in the Sloan Digital Sky Survey, only twelve candidates were identified, and just three confirmed. These systems are exceptionally valuable because they allow astronomers to precisely measure the mass of a quasar's host galaxy, something that is normally impossible given that the overwhelming brightness of the quasar itself drowns out its surroundings.
Now researchers led by Everett McArthur have dramatically expanded this tiny sample using an innovative machine learning approach and data from the Dark Energy Spectroscopic Instrument.
Their study, published on the arXiv preprint server, examined over 812,000 quasars and identified seven new high quality candidates, more than doubling the known sample in a single search.
The challenge lies in detecting the subtle signature of a background galaxy whose light has been gravitationally bent by the quasar's host galaxy in the foreground.
When a more distant galaxy sits almost perfectly aligned behind a quasar, the immense gravity of the quasar's host galaxy acts as a lens, bending the background galaxy's light around it. This produces multiple distorted images of the background source, though these are typically too faint and small to resolve from the ground given the quasar's brilliant glare.
Spectroscopy offers a different detection method. If a background galaxy's light passes through the same spectrograph fiber as the foreground quasar, its emission lines appear at a different wavelength due to its higher redshift. The researchers trained a neural network to spot these telltale features buried in quasar spectra.
Since genuine quasar lenses are extremely rare, the team couldn't train their neural network on thousands of real examples. Instead, they constructed realistic mock lenses by combining actual DESI spectra of quasars with spectra of higher redshift emission line galaxies.
They fed approximately 3,000 synthetic lenses and 30,000 ordinary quasar spectra into the network, teaching it to distinguish the subtle emission line signatures of background galaxies from the complex spectral features of quasars themselves. The network achieved a classification performance with an area under the curve of 0.99, exceptionally high accuracy.
Applying this approach to DESI's first data release, which spans quasars at redshifts between 0.03 and 1.8, they identified seven Grade A candidates. Each shows a strong oxygen doublet emission line at a higher redshift than the foreground quasar, and four additionally display hydrogen beta and oxygen three emission from the background galaxy.
The method even successfully recovered the single previously known quasar lens system that fell within DESI's footprint.
Why does this matter? Quasar lenses provide a powerful probe of how supermassive black holes and their host galaxies co-evolved across the history of the universe. The Einstein radius (the characteristic angular size of the lensed images) directly reveals the host galaxy's mass.
With traditional methods, teasing apart the quasar's light from its host galaxy is nearly impossible, but gravitational lensing makes this measurement straightforward.
+------------------------------------------------------------------------------+
================================================================================
= ____ __ ____ ___ _ _ ____ ____ ____ ____ =
= ( __)( )(___ \ / __)( \/ )( _ \ ( _ \( _ \/ ___) =
= ) _) )( / __/( (_ \ ) / ) _ ( ) _ ( ) _ (\___ \ =
= (____)(__)(____) \___/(__/ (____/ (____/(____/(____/ =
= Serving The Irish Packet Radio Network Since 2006 =
= Packet: EI2GYB@EI2GYB.DGL.IRL.EURO / EI2GYB@WINLINK.ORG =
= Email/PayPal: EI2GYB@GMAIL.COM =
================================================================================
Read previous mail | Read next mail